University of Nebraska - Lincoln DigitalCommons@University of Nebraska - Lincoln

Great Plains Research: A Journal of Natural and Social Sciences Great Plains Studies, Center for

August 1991

Frontier Flintlocks: A Fault Tree Analysis of Firearm Use at Contact Period Sites of the Great Plains

Peter Bleed University of Nebraska - Lincoln, [email protected]

Daniel Watson University of Nebraska - Lincoln

Follow this and additional works at: https://digitalcommons.unl.edu/greatplainsresearch

Part of the Other International and Area Studies Commons

Bleed, Peter and Watson, Daniel, "Frontier Flintlocks: A Fault Tree Analysis of Firearm Use at Contact Period Sites of the Great Plains" (1991). Great Plains Research: A Journal of Natural and Social Sciences. 25. https://digitalcommons.unl.edu/greatplainsresearch/25

This Article is brought to you for free and open access by the Great Plains Studies, Center for at DigitalCommons@University of Nebraska - Lincoln. It has been accepted for inclusion in Great Plains Research: A Journal of Natural and Social Sciences by an authorized administrator of DigitalCommons@University of Nebraska - Lincoln. FRONTIER FLINTLOCKS: A FAULT TREE ANALYSIS OF FIREARM USE AT CONTACT PERIOD SITES OF THE GREAT PLAINS

Peter Bleed and Daniel Watson

Department of Anthropology University of Nebraska Lincoln, NE 68588-0368

Abstract. Gun part assemblages from several Euroamerican and Native American contact period sites from the Plains are compared as a way of examining how firearms were incorporated into Native technology of the Plains region. These data are interpreted in terms of a 'tfault tree analysis," an operations research technique that identifies potential points of failure in technical systems in order to study patterns of use, maintenance, and reliability. The analysis indicates distinctively different patterns of gun repair and treatment by Indians and Euroamericans but suggests that Indians were quite capable of repairing firearms and that they systentatically reused parts from failed arms.

The period of initial contact between Euroamerican and Native American societies has attracted the attention of historians, anthropolo- gists, archaeologists, and, of course, the general public. Beyond their intrinsic interest, the dramatic events and romantic objects of the contact period have been studied because they were pivotal to the subsequent history and cultural developments of North America (Billington 1967). As a period of rapid acculturation, the contact period also provides an opportunity to study what happens when radically different cultures are thrown into contact (White 1975). Contact period sites have attracted the attention of archaeologists and excavations of both Indian and Euroamerican sites have contributed substantially to understanding how the frontier progressed. Archaeological research has shed specific light on the demographic implications of Euroamerican contacts (Ramenovsky 1988) and has shown how materials flowed through the cultural systems that developed as Euroamericans invaded the continent. Detailed analyses of trade goods as diverse as pipe stems (Binford 1961) and buttons (Olsen 1963) have added fine chronolog- ical control to investigation of the frontier. Other studies have documented the material culture of frontier societies (Quimby 1966) and isolated distinct ethnic and national spheres of influence (South 1978). Moving beyond descriptive concerns, archaeologists addressed the processual study of the culture change that marked the frontier (Lewis 234 Great Plains Research Vol.1 No.2

1984). At first, contributions in this regard consisted of giving substance to theories drawn from other fields (Ray 1978) or articulating written and material records (Pyszczyk 1989). Increasingly, archaeologists aim at using material records to study the cultural interactions that marked the frontier (Brown 1979; Green 1985). This paper addresses acculturation that occurred on the frontier in strictly archaeological terms. It considers the use and discard patterns of flintlock firearm parts to discuss how Euro- american goods were accommodated by Indian technology. Flintlocks have been one of the classic icons of frontier period archaeology and are an appropriate focus for consideration of technologi- cal acculturation for several reasons. They have received a great deal of archaeological attention because of their interest to collectors and because they are sensitive chronological and cultural markers. Flintlock firearms were popular and important trade goods, used by Euroamericans and much sought after by Indian hunters and warriors. Historic and ethnographic studies have shown that, after an initially slow accommodation (Babits 1976), flintlock arms were easily incorporated into native economic and social patterns (Secoy 1953). Firearms were, thus, responsible for intensifications of preexisting social and subsistence patterns where they could be accepted without requiring entirely new cultural patterns. The way firearms were incorporated into Indian technology is less well documented in historical sources, which contain scattered anecdotal references but little systematic information on how Indians handled and treated the guns they obtained through trade. Technologically, flintlock arms were exotic to Indian communities and represented a dramatic break with tradition. Firearms incorporated materials and techniques unknown to stone-age cultures and required maintenance patterns and tools that were very different from those of traditional Indian weapons. If Indian technological adjustment to firearms is hard to address through historic sources, it can be studied archaeologically because the archaeological record preserves the direct residues and results of gun handling. Moreover, since data are available from both Euroamerican and Indian sites, there is information available on flintlock usage on both sides of the frontier. In this paper, archaeological assemblages of gun parts from contact period Indian and Euroamerican sites of the Great Plains are compared to gain insights on how firearms were incorporated into Indian technologies.

Fault Tree Analysis

To organize the comparisons and focus on the processes that were responsible for the archaeological assemblages considered here, we use Frontier Flintlocks 235

All immediate subsequent events -- IIAND" Gate must occur for system to fail.

Any immediate subsequent event -- OR" Gate can cause system to fail.

A fault that can be -- FAILURE further decomposed.

fi A fault that cannot be )-- BASICFAILURE further decomposed.

A fault purposely UNDEVELOPED -- not developed. FAILURE

Figure 1. Standard fault tree symbols.

fault tree analysis. This is an operations research technique developed by engineers to determine the array of adverse conditions--or faults-that can cause a technological system to fail (see McNitt 1986; Henley and Kumamoto 1981). Fault trees are dendritic models that present a task's structure by showing how sequences of actions are interconnected. Fault trees focus on the things that can go wrong during the operation of the task. The central element of analysis is a fault tree model, which lays out the conditions that can result in a system's failure. Virtually any technical or social system can be treated in this way. Fault trees are hierarchical in that they lay out several levels of potential failure. The diagram tree starts with a basic system failure. This "top event" is then decomposed into all of the faults (subevents) that can contribute to it. On the diagram, the alternative faults that can possibly beset a system are logically presented and related to one another and to the top event by means of a series of conventional symbols (Fig. 1). Below the top event, potential intermediate 236 Great Plains Research Vol.1 No.2 failure events--or those that can be further decomposed into the subevents -are symbolized by rectangles. "Basic failure events," presented in circles, are faults that cannot be further decomposed. They form the bottom of the tree. Diamond shapes indicate failure events that could be further disaggregated but which are not considered in the analysis. Two types of "gates" connect the various kinds of events: "AND" gates connect subevents that can cause failure only when they all occur together; "OR gates are located between subevents that may independently cause a failure. A fault tree for a complex system with multiple subsystems or built-in redundancies can become complex. The purpose is, however, to lay out in an interpretable order the range of problems a system might encounter. Applied to archaeological materials, a fault tree offers a means of conceptually linking failed residues in the archaeological record to the operation of larger behavioral systems and technical activities (Bleed 1991).

The Flintlock Fault Tree

A standard flintlock firearm is a relatively simple system that can easily be presented as a fault tree (Fig. 2). Most flintlocks consist of three subsystems-the lock, stock, and barrel-that operate in parallel and without backups. If any one of these fail, the gun will not fire. The flintlock fault tree, therefore, starts with a single "OR" gate indicating that a loaded flintlock gun can be inoperable due to failure in one of its subsystems. The barrel fails through explosion, excessive wear, or if the touch hole at the breech is enlarged. In general, an exploded barrel reflects either misuse of a gun or a manufacturing defect. Worn ("shot out") barrels and enlarged touch holes are results of normal, but very long, use. The wooden stock holds the subsystems together in proper alignment and makes the whole arm manageable. If it breaks through misuse the arm is inoperable. Metal fittings strengthen and decorate the stock, but most are not critical to the operation of the arm. The lock is the most delicate and complex subsystem within a flintlock and the fault tree clearly shows that it is the subsystem with the most potential for failure (Fig. 3). A worn flint that cannot generate a spark presents a basic failure, but this is one of a number of maintenance problems that can be set right by the operator. Other lock faults require more extensive repairs or adjustments. The cock can break off the gun. The frizzen can be excessively worn so that it will not strike a spark. In some designs, the pan can be separated from the lock plate. The sear can be worn so that the cock is not controllable. Finally, failure of any of the three springs that control lock parts can be considered a basic fault. Frontier Flintlocks

SHOOT

Figure 2. A flintlock fault tree.

By indicating what can go wrong with a flintlock, the fault tree presents a set of expectations about what residues should normally result from flintlock use. Since failure of any subsystem causes total failure, the archeological record should include lock, stock, and barrel parts. Still, the model shows that the lock is the subsystem most prone to failure since it has a large number of critical parts that must operate in precise articula- tion with one another. Great Plains Research Vol.1 No.2

Figure 3. External (top) and internal (bottom) views of a flintlock showing parts subject to failure: D. Flint, B. Cock, K. Frizzen, G. Pan, I. Sear, L., M. and H. Springs. (From Gillispie. 1959: Plate 61). Frontier Flintlocks 239

Flintlocks on the Frontier

To examine the practical realities of gun use on the frontier, published and unpublished data on gun part assemblages from early historic period Euroamerican and Indian sites on the Great Plains were gathered (Table 1, Fig. 4). Contact period assemblages from eastern and southwestern North America and other areas were not considered because they would introduce a wide range of historic and cultural variables. Limiting the scope of the study to the central and northern Plains focuses analysis on an interval and area that experienced a limited range of influences and trends. To emphasize how guns were used, only materials from domestic residential areas were initially considered. Burial assemblages were not included in the study sample for two reasons. First, culture historic research questions that have dominated Plains research for the past 60 years have made Indian burials more interesting that those of Euro- americans. Data available on non-Indian burials are so limited and widely scattered as to be incomparable to the relatively large body of data available on Indian burials. Second, it seem probable that firearms included with burials reflect less on gun use than on other social patterns. As data collection progressed, however, some burial assemblages clearly became significant in understanding Indian patterns of flintlock usage and are discussed below. Only the gun parts contained in the collections were considered. No attempt was made to identify the types of guns reflected in the collections. Generally, the sites yielded only flintlock remains, but all of the gun parts present in the assemblages were considered. Contact period sites that yielded no gun parts are, of course, not reflected in the sample. Finally, while gathering data, the focus was on site assemblages. Finer chronologi- cal or provenance subdivisions were not considered. Gun part assemblages from 18 Indian and seven Euroamerican community sites differed markedly (Table 2). Since the fault tree model indicated that potential for flintlock failure is concentrated in the lock, it is not surprising that lock elements account for nearly 40% of the gun parts from Euroamerican domestic sites. The relatively low proportion-- about one quarter--of failed lock parts from Indian communities is remarkable. Likewise, the differences in the frequency of barrel fragments is very notable. They are rare in Euroamerican sites, but comprise more than half of the gun parts from the Indian communities. A chi-square test of the distribution of parts in the domestic site assemblages substantiates that the distribution is nonrandom (x2=76.65, df=2, p= >.0001). A broader perspective on the domestic site data is provided by three 240 Great Plains Research Vol.1 No.2

TABLE I ASSEMBLAGES CONSIDERED IN THIS ANALYSIS

Site Affiliation Date # of Reference parts or location

Native American Domestic Sites Biesterfeld Cheyenne ca 1700 1 Wood 1971 32ML2 Hidatsa el al. 1845-1870s 49 Smith 1972 32ME12 Hidatsa et al. pre-1844 1 Ahler & Swenson 1985 320L110 DakotatArikara 1831-1855? 7 Smith 1986 32ME5 Mandan 1787-1822 32 Lehmer et al. 1978 32ME8 Ari ka ra 1798-1834 6 Lehmer et al. 1978 32ME15 Arikara ca 1800 2 Lehmer el al. 1978 39BF2 Arikara mid-1700s 4 Ahler & Toom 1989 39ST6 Arikara 1740-1795 1 Lehmer & Jones 1968 39C09 Ari kara 1800-1823 17 Krause 1972 25NC2 Pawnee 1809-1844 16 NSHS* 25PK1 Pawnee 1823-1846 9 NSHS 25SD2 Pawnee 1852-1859 6 NSHS 25SD1 Pawnee ? 9 NSHS 25GA1 Pawnee 1809? 1 NSHS 25BU1 Pawnee 1750-1809 8 NSHS 25HW1 Pawnee 1770-1844 3 NSHS 25WT1 Pawnee 1775-1809 14 NSHS 25NC7 Pawnee 1842-1847 2 NSHS 25BU4 Pawnee 1750-1770s 7 NSHS 14RP1 Pawnee 1770s 19 Roberts 1978 23SA3 Osage 179-1775 100 Hamilton 1982a 23VE4 Osage 1775-1815 12 Hamilton 1982a 23VE3 Osage 1775 45 Hamilton 1982a 23VE1 Osage 1790-1815 247 Hamilton 1982a Euroamerican 32WI17 Ft. Union 1829-1867 116 Hunt 1986:125-129 32MN1 Kipp's Post 1820s 1 Woolworth & Wood 1960 32ML2Ft Berthold I 1845-1862 35 Smith 1972 32ML2 Ft. Berthold I1 1851-1870s 17 Smith 1972 39ST217 Ft. Pierre 1859-1863 8 Caldwell 1982 25 WN9 Ft. Atkinson 1820-1827 14 NSHS 25SY26 Fontenelle's Post 1822-1842 18 NSHS Cachesburials 25DK2 Omaha ca 1810 115 UNL** 25PK1 Pawnee early 1800s 23 NSHS 23SA3a Osage ca 1750 108 Hamilton 1982b

*** Nebraska State Historical Society University of Nebraska-Lincoln, Department of Anthropology Frontier Flintlocks 24 1

Figure 4. Sites considered in this analysis.

caches of gun parts made by Indians away from their communities. Two of these are from burials, one Pawnee (25PK1) and one Omaha (25DK2). The third is a large cache of parts and tools found in an isolated spot less than one mile from the Little Osage village (23SA3) in western Missouri (Hamilton 1982b). This cache can be assumed to have been made by an Indian, probably an Osage, because of its proximity to the Osage village. These caches obviously form a small sample, but they markedly amplify the Indian gun part assemblage recovered from domestic sites. The cache assemblages are overwhelmingly composed of lock parts, precisely the parts that appear underrepresented in the domestic assemblages. When Great Plains Research Vol.1 No.2

TABLE 2

COMPARISON OF GUN PART ASSEMBLAGES FROM INDIAN AND EUROAMERICAN RESIDENTIAL COMMUNITIES

Communities Lock Stock Barrel Total

Indian 167 154 297 618

Euroamerican 8 1 98 30 209

the cache assemblages are combined with the domestic totals, the assemblage of gun parts from Indian and Euroamerican sites become more similar (see Table 3). With the cached parts included in the Indian total, barrels and barrel fragments continue to form a relatively large part of Indian assemblages and stock elements still remain relatively rarer than in Euroamerican assemblages. The overall distribution of arts, thus, still shows a statistically significant non-random distribution (Xe=62.818. df=2, p=>.0001). In terms of the proportion of lock parts, however, the assemblages have become markedly similar.

Discussion and Conclusions

The gun part assemblages considered here were unquestionably influenced by factors ranging from the design of flintlock guns to the research strategies of Plains archaeologists. In part, at least, the differences between assemblages from Indian and Euroamerican sites must also reflect that firearms were treated differently by Indians and Euroamericans on the Great Plains frontier. The nature and significance of those differences, however, appear to have been anything but simple. Gun part assemblages from Indian residential sites include a preponderance of barrels. This high frequency may in part reflect the fact that light-barreled fusils were popular among Indians. These barrels may have had a shorter lifespan than heavier rifle barrels preferred by Euroamericans (Hamilton 1980:7-8). Beyond that potential, many of the barrel fragments found in Indian communities had been reworked into Frontier Flintlocks

TABLE 3

COMPARISON OF GUN PART ASSEMBLAGES FROM INDIAN AND EUROAMERICAN DOMESTIC SITES AND CACHES

Communities Lock Stock Barrel Total

All Indian assemblages 400 193 298 89 1

(n =28) (44.9%) (21.7%) (33.4%) (100.0%) .

Euroamerican 8 1 98 30 209

(n=7) (38.8%) (46.9%) (14.3%) (100.0%)

scrapers or other tools. Thus, the high frequency of barrel remnants in Indian sites also probably indicates that they had relatively high scrap value to Indians who used them in ways that allowed them to become part of the domestic residues. In contrast to Euroamerican assemblages, the relatively low proportion of stock parts apparent in Indian domestic assemblages may further reflect regular recycling of the parts of failed guns. Small stock futures of brass or other metals turned into ornaments and small tools would have been highly portable. Unlike bulky barrel fragments, such small pieces may have been kept out of the archaeological record, and the study sample, because they could easily be lost or discarded in areas away from residential communities. The mix of gun parts from Indian communities would also suggest that guns were only irregularly repaired in them. Aside from the prepon- derance of barrels, Indian community assemblages present a rather even mix of parts from all of the flintlock subsystems. This kind of mix would be consistent with a pattern of rare or irregular gun repair since, if specific faults were not repaired, any failure would result in total discard of the entire arm. In this situation all kinds of parts would appear in the archaeological record. Regular repair, on the other hand, would cause a relatively high archaeological frequency of the specific parts most prone to failure and of those parts least worth preserving for reuse. Euroamerican sites have yielded the kinds of assemblages consistent with regular firearm repair. Systematic lock repair by individuals expert 244 Great Plains Research Vol.1 No.2 enough to undertake specific technical repairs of failed systems would result in assemblages like those from Euroamerican sites. The rarity of barrels in those site may further reflect lock repair since a well-maintained, high quality barrel could outlast several locks. Furthermore, with access to blacksmiths, Euroamericans may have been able to recycle barrels as scrap iron or in some other assertive ways that made them unrecognizable in the archaeological record. The high frequency of stock fittings in Euroamerican communities presumably indicates that these parts were of little value once the gun they were on had failed beyond repair. They were not subject to curation and appear instead to have been treated like entirely valueless discardables. Given only the assemblages from domestic sites, then, it would be tempting to conclude that regular maintenance was more common among Euroamerican frontiersmen than Indians. That conclusion would fit the intuitive expectation that Indians may have had trouble accepting unfamiliar mechanical systems like those of the flintlock. With easier access to both technical experts and spare parts, minor repairs must have been easier in Euroamerican communities where a failed lock could be repaired or still usable parts (like barrels) fitted to other guns and, therefore, kept out of the archaeological record. Domestic site data would appear to indicate that goods spread more rapidly than technical knowl- edge on the frontier and that technical expertise developed slowly in Indian communities. Those simple conclusions, however, are cast into considerable doubt by the contents of gun part caches made by Indians away from their residential areas. When the contents of the three known caches are combined with the community assemblages, Indian patterns of flintlock use appear both complex and sophisticated. The high frequency of reused barrel fragments, and perhaps even the relative rarity of stock fixtures, can be seen as evidence that guns parts were highly valued and systematically recycled. Even after they became useless as weapons, Indians appear to have treated their guns as valuable technological resources. Furthermore, when the caches are included, the similarity in the proportion of lock parts on Indian and Euroamerican sides of the frontier certainly suggests that even during the contact period Indians were quite capable of systematically repairing even the most technically complex and delicate parts of their guns. The fact that the caches represent three different tribes may indicate that the patterns of curation and repair they reflect were widespread among Plains groups. Archaeology does reveal differences between Indian and Euro- american patterns of flintlock treatment but those differences do not appear to reflect significantly different levels of skill or understanding. Rather, differential access to resources and different social arrangements Frontier Flintlocks 245 are more likely explanations for the variation. Parts from failed guns appear to have been relatively more valuable in Indian communities so that they were systematically recycled and reused. The relatively high value of recycled gun parts may actually have inhibited repairs in Indian communities. Furthermore, if the parts caches from burials were goods interred as the special property of the individuals who had assembled and sorted them, the caches would appear to indicate that technical specialists had begun to develop in Indian communities. Finally, this research offers two methodological lessons that are worth noting. First, by focusing on the operating processes of technology, fault tree analysis provides an interesting framework for gathering and interpreting archaeological data. Second, the project offers a substantive illustration of the potential importance that archaeological information from burials can have for the full understanding of the human past.

Acknowledgrnen ts

This project has profited from the expertise and support of many individuals. John Ludwickson, Rob Bozell, and Gayle Carlson freely shared their expertise in our examination of the Nebraska State Historical Society collection. Bill Hunt and Virgil Noble of the National Park Service, Midwest Archaeological Center provided assistance with their data. James Hanson, Warren Caldwell, John Ludwickson, and Bill Hunt offered useful comments on early drafts of this paper. We appreciate their assistance, but emphasize that no one but ourselves should be held accountable for failings that remain. We presented a version of this paper at the 1990 Annual Meeting of the Society for American Archaeology.

References

Ahler, S. and D. Toom, eds. 1989. Archeology of the Medicine Crow Site Complex Bufialo County, South Dakota. Springfield, IL: Illinois State Museum Society submitted to the Branch of Interagency Archeologi- cal Services, Denver, CO: US National Park Service, Rocky Mountain Regional Office. Ahler, S. and A. A. Swenson 1985. Test Excavations at Big Hidatsa Village (32ME12) Knife River Indian villages National Historic Sire. Grand Forks, ND: Department of Anthropology, University of North Dakota. Submitted to the Midwest Archeological Center, Lincoln, NE: US National Park Service. Babits, L. E. 1976. The evolution and adoption of firearm ignition systems in eastern North America: An ethnohistorical approach. The 246 Great Plains Research Vol.1 No.2

Chesopiean: A Journal of North American Archeology 14:June-August. Billington, R. A. 1967. The American frontier. In Beyond the Frontier: Social Process and Cultural Change, ed. P. Bohannon and F. Plog, 3-24. Garden City, NJ: Natural History Press. Binford, L. 1961. A new method of calculating dates from kaolin pipe stems. southeastern Archaeological Conference Newsletter 9 No. 1. Bleed, P. 1991. Operations research and archaeology. Anterican Antiquity 56: 19-35. Brown, I. W. 1979. Functional group changes and acculturation: A case study of the French and the Indian in the lower Mississippi Valley. Mid-Continental Journal ofArchaeology 4147-65. Caldwell, W. W. 1982. Relics from Fort Pierre 11, Oahe Reservoir, South Dakota. In Indian Trade Guns, ed. T. M. Hamilton, 115-26 (reprint of Missouri Archaeologist Vol. 20, 1960) Union City, TN: Pioneer Press. Gillispie, C. G. ed. 1959. A Diderot Pictorial Encyclopedia of Trades and Industry. Dover, NY: Dover Publications. Green, S. W. and S. M. Perlman. 1985. The Archaeology of Frontiers and Boundaries. New York, NY: Academic Press. Hamilton, T. M. 1980. Colonial Frontier Guns. Chadron, NE: The Fur Press. Hamilton, T. M. 1982a. Relics from 18th century Osage sites. In Indian Trade Guns, ed. T. M. Hamilton, 71-105. (reprint of Missouri Archaeologist 20:1960) Union City, TN: Pioneer Press. Hamilton, T. M. 1982b. The gunsmith's cache from Malla Bend, Missouri. In Indian Trade Guns, ed. T. M. Hamilton, 29-52. (reprint of Missouri Archaeologist 20:1960) Union City, TN: Pioneer Press. Henley, E. T. and H. Kumamoto. 1981. Reliability Engineering and Risk Assessment. Englewood Cliffs, NJ: Prentice Hall. Hunt, W. 1986. Fort Union Trading Post National Historic Site (32WI17) material culture report, part IV: Firearms, trapping and fishing equipment. Report on file Lincoln, NE: US National Park Service, Midwest Archaeological Center. Krause, R. 1972. The Leavenworth Site: Archaeology of An Historic Ankara Communiry. Lawrence, KS: Publications in Anthropology, University of Kansas. Lehmer, D. J. and D. Jones. 1968. Ariknra Archeology: The Bad River Phase. Lincoln, NE: Smithsonian Institutions River Basin Surveys, Publications in Salvage Archeology, No. 7. Lehmer, D., W. R. Wood, and C. L. Dill. 1978. The Knife River Phase. Denver, CO: The Interagency Archeological Service, fulfillment of Contract C3537(68). Lewis, K. 1984. The American Frontier. New York, NY: Academic Press. Frontier Flintlocks 247

McNitt, R. P. 1986. Fault tree techniques. In Failure Analysis in Microsco- py, Fractography and Failure Analysis, ed. M. R. Louthan and T. A. Place, VII, 1-10. Blacksbung, VA: International Metallographic Society, American Society for Metal, Failure Analysis and Prevention Laboratory of Virginia Polytechnic Institute. Olsen, S. J. 1963. Dating Earl plain buttons by their form. Anterican Antiquity 28551-54. Pyszczyk, H. 1989. Consumption and ethnicity: An example from the fur trade in western Canada. Journal of Anthropological Archaeology 8:213-49. Quimby, G. 1966. Indian Culture and European Trade Goods, the Archaeol- ogy of the Historic Period in the Western Great Lakes Region. Madison, WI: University of Wisconsin Press. Ramenovsky, A. F. 1988. Vectors of Death: The Archaeology of European Contact. Albuquerque, NM: University of New Mexico Press. Ray, A. 1978. History and archaeology of the northern fur trade. Anlerican Antiquity 34:26-34. Roberts, R. L. 1978. The archaeology of the Kansas monument site: A study in historical archaeology of the Great Plains. M. A. Thesis, University of Kansas. Secoy, F. R. 1953. Changing Militaly Patterns on the Great Plains. Locust Valley, NY: August in. Smith, C. S. 1972a. Firearms, gunflints, ammunition and military gear from Like-a-Fishhole Village. In Like-a-Fishhook Village and Fort Berthold, Garrison Reservoir, North Dakota, ed. G. H. Smith, 80-88. Washing- ton, DC: US National Park Service, Anthropological Paper No. 2. Smith, C. S. 1972b. Firearms, ammunition and military gear from Fort Berthold I. In Like-a-Fishhook Village and Fort Berrhold, Garrison Reservoir, North Dakota, ed. G. H. Smith, 108-10. Washington, DC: US National Park Service, Anthropological Paper No. 2. Smith, C. S. 1972c. Firearms, gunflints, and ammunition from Fort Berthold 11. In Like-a-Fishhook Village and Fort Berthold, Gam'son Reservoir, North Dakota, ed. G. H. Smith, 163-65. Washington, DC: US National Park Service, Anthropological Paper No. 2. Smith, C. S. 1986. Euro-American weapons. In Papers In Northern Plains Prehistoy and Ethnohistoly: Ice Glider-320LllO. Sioux Falls, SD: Special Publication of the South Dakota Archaeological, No. 10. South, S. 1978. Pattern recognition in historic archaeology. Anlerican Antiquity 43:223-30. White, J. R. 1975. Historic contact sites as laboratories for the study of culture change. In The Conference on Historic Site Archaeology Papers, ed. S. South. Vol. 9, 153-63. Columbia, SC: Institute of Archeology 248 Great Plains Research Vol.1 No.2

and Anthropology, University of South Carolina. Wood, W. R. 1971. Biesterfeldt: A Post Contact Coalescent Site on the Northeastern Plains Contribution to Anthropology No. 15. Washington, DC: Smithsonian Institution Press. Woolworth, A. and W. R. Wood. 1960. The archaeology of Kipp's Post (32MN1), River Basin Survey Paper #2,239-305. Bureau ofAmerican Ethnology, No. 176. Washington, DC: Smithsonian Institution.